patent-management-system
Pass
Audited by Gen Agent Trust Hub on Aug 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill generates HTML that references
https://cdn.jsdelivr.net/npm/chart.js@4.4.0/dist/chart.umd.min.js. This is a well-known content delivery network (CDN) used for legitimate data visualization purposes. - [INDIRECT_PROMPT_INJECTION]: The skill exhibits an attack surface where instructions or data from external patent databases could influence the generated output.
- Ingestion points: Patent data is ingested via the
patsnap_searchtool as described inSKILL.md. - Boundary markers: The instructions do not specify the use of delimiters or 'ignore' instructions when interpolating API data into the HTML template.
- Capability inventory: The skill uses
patsnap_searchfor data retrieval and instructs the agent to use Python file tools to write the resulting HTML to the local filesystem. - Sanitization: There are no explicit instructions to sanitize or escape the patent titles, summaries, or other metadata before they are embedded into the generated HTML, which could lead to cross-site scripting (XSS) if the source data is malicious.
Audit Metadata